Malnutrition was significantly correlated with HBI (p < 0,002), CRP (p < 0,032) and presence of inflammation at the endoscopy (p < 0,05). Amongst the 5 GLIM criteria, % of WL, low BMI, strength diminution, decreased oral intake and active inflammation were defined as positive in 43,33%, 23,3%, 76,7%, 56,7% and 83,3% of cases respectively. SGA and GLIM criteria were positively correlated in 63% of the patients. Conclusions: Based on the GLIM criteria, the prevalence of malnutri- tion was 32,6% in a cohort of ambulatory CD patients. Amongst the 5 criteria, strength, decreased oral intake and the inflammation were the most frequently altered. Malnutrition was correlated with the clinical, the biological and the endoscopic activity of the disease. Disclosure of Interest: None declared. SUN-PO225 QUANTITATIVE ANALYSIS OF NUTRITIONAL RISK SCREENING TOOLS NRS 2002 AND PG-SGA IN ONCOLOGYOUTPATIENTS E. Justin 1* , D. Mlakar Mastnak 1 , N. Rotovnik Kozjek 1 . 1 Unit of Clinical Nutrition and Diet Therapy, Institute of Oncology Ljubljana, Ljubljana, Slovenia * Corresponding author. Rationale: In the clinical pathway of nutritional care for hospitalized patients at the Institute of Oncology Ljubljana Nutrition Risk Screening NRS 2002 (NRS 2002) has already been implemented. Now, choosing an effective tool for nutritional screening for patients treated in our outpatients department also becomes a key objective. The purpose of the study was to evaluate the use of NRS 2002 and Patient-Generated Subjective Global Assessment (PG-SGA) in outpatient department. The goal was to determine the differences in the prevalence of patients at nutritional risk (NR) and the usefulness of the two tools. Methods: In one week in the 2013,144 patients were included. The patients were interviewed by a dietitian using NRS 2002 and PG-SGA. In terms of the number of points acquired, patients were classified into three categories: not at NR (PG-SGA: 0–3; NRS 2002: 0); at moderate NR (PG-SGA: 4–8; NRS 2002: 1–2); at high NR (PG-SGA: ≥9; NRS 2002: ≥3). The data were analyzed with the statistical program SPSS 22.0, we used the t-test. Results: The difference in prevalence of patients at NR was 9,6% (p < 0,05). The prevalence by using PG-SGA was 56,2% and by NRS 2002 46,6%. PG-SGA detected 6,9% more patient at moderate NR than NRS 2002 (PG-SGA: 24,3%, NRS 2002: 17,4%). Patients at high NR were about one third, the difference between tools was smaller (PG-SGA: 31,9%, NRS 2002: 29,2%). Conclusions: Although PG-SGA detected more patients at NR and included nutritional assessment and triage for nutritional support, NRS 2002 was less time-consuming and easyer to perform as such was more suitable for use in our outpatient clinics. On the basis of the obtained results, we suggest the use of NRS 2002, but we recommend that nutritional support should be performed as soon as patient is at nutritional risk (NRS 2002 1-2 points). Disclosure of Interest: None declared. SUN-PO226 NUTRITIONAL RISK SCREENING IN ONCOLOGY OUTPATIENTS: SF PG-SGA VERSUS PG-SGA AND NRS 2002 E. Justin 1* , D. Mlakar Mastnak 1 , N. Rotovnik Kozjek 1 . 1 Unit of Clinical Nutrition and Diet Therapy, Institute of Oncology Ljubljana, Ljubljana, Slovenia * Corresponding author. Rationale: Short Form Patient-Generated Subjective Global Assessment (SF PG-SGA) is a validated tool for nutritional risk screening (NRS). Beside being shorter than PG-SGA, it can be filled by the patient himself. This is particularly important in situaton with a large number of patients in our outpatient department and lack of medical staff. The purpose of the study was to evaluate the use of (SF PG-SGA), Patient-Generated Substantive Global Assessment (PG-SGA) and Nutritional Risk Screening 2002 (NRS 2002) in the outpatient department of the Institute of Oncology Ljubljana. The goal was to determine the differences in the prevalence of patients with nutritional risk (NR) using all three tools. Methods: In one week in the 2013, 144 patients were included. In terms of the number of points acquired in NRS with all three tools, patients were classified into three categories: not at NR (SF PG-SGA: 0– 2, PG-SGA: 0–3, NRS 2002: 0), at moderate NR (SF PG-SGA: 3–5, PG- SGA: 4–8, NRS 2002: 1–2) and at high NR (SF PG-SGA: ≥6, PG-SGA: ≥9, NRS 2002: ≥3). Results: The prevalence of NR patients using SF PG-SGAwas 49,3% (at moderate NR 18,0%, at high NR: 31,3%), using the PG-SGA 56,2% (at moderateNR: 24,3%, at high NR: 31,9%), using the NRS 2002 46,6% (at moderate NR: 17,4%, at high NR: 29,2%). Not at NR patients were 50,7% using the SF PG-SGA, 43,8% using the PG-SGA and 53,4% using the NRS 2002. Between SF PG-SGA and PG-SGA, the difference in prevalence was 6,9%, and between SF PG-SGA and NRS 2002 it was 2,7%. Conclusions: The difference in the prevalence of NR patients between SF PG-SGA and NRS 2002 was 2,7% especially it was small in the moderate NR category (0,6%). But SF PG-SGA and NRS 2002 compared with PG-SGA detected fewer all at NR patients and not included nutritional assessment. However, SF PG-SGA in comparison with NRS 2002 may have an advantage over NRS 2002 because it includes the assessment of patients nutritional symptoms and their physical condition. Disclosure of Interest: None declared. SUN-PO227 CONCORDANCE OF DIFFERENT TOOLS FOR DIAGNOSIS OF MALNUTRITION IN HOSPITALIZED PATIENTS F.M. Silva 1* , C.F. Burgel 1 , B.E. De Araújo 1 , P. Teixeira 1 , F.W. Rodrigues 1 , V. Chites 1 , G.D. Carvalho 2 , C.B. Saragiotto 3 , P.V.G. Modanesi 4 , E. Rabito 5 . 1 nutrition, Universidade Federal De Ciencias Da Saude De Porto Alegre, Porto Alegre, 2 nutrition, Universidade Federal do Paraná, 3 nutrition, UFPR, 4 Hospital of Clinics – UFPR – EBSERH, 5 UFPR, Curitiba, Brazil * Corresponding author. Rationale: There is no international consensus on a single ‘best tool’ for malnutrition diagnosis, and new tools were proposed by different international society of clinical nutrition recently (1–3). Considering that the concordance of these tools with Subjective Global Assessment (SGA) has been scarcely investigated until this moment, the aim of this study was to evaluate the concordance of different tools for diagnosis of malnutrition in hospitalized patients with SGA. Methods: Cross-sectional study including Brazilian patients from two hospitals (Porto Alegre-RS and Curitiba-PR). The inclusion criteria were age ≥18 years, conscious and able to move patients. Malnutrition was diagnosed within 48 hours of hospitalization by SGA, Academy of Nutrition and Dietetics/American Society for Parenteral and Enteral Nutrition (AND/ASPEN), European Society for Clinical Nutrition and Metabolism (ESPEN), and Global Leadership International Malnutrition (GLIM) criteria. The kappa coefficient was calculated and interpreted as follow: 0.41–0.60 moderate; 0.61–0.80 substantial; 0.81–1.0 almost perfect. Results: 509 patients (50.3 ± 14.7 years old; 51.5% females; 52.3% from Curitiba) were studied. The majority of patients had cancer (25.9%), cardiovascular disease (19.6%) and gastrointestinal disorder (14.5%). Malnutrition was identified in 33.2% of patients by SGA, 36.4% by AND/ASPEN, 16.1% by ESPEN and 37.5% by GLIM. The concordance between SGA and AND/ASPEN (k = 0.66; p < 0.001) and GLIM (k = 0.69; S143 Abstracts / Clinical Nutrition 38 (2019) S59–S296